From education to health and nutrition to natural resource management, Tata Trusts’ footprint in a wide array of sectors makes it one of the most authentic voices on complex development issues in India. Now the Mumbai-headquartered organization is throwing its might behind a crucial reform process: championing data-driven governance to ensure that the state’s policies can generate the maximum developmental impact on the ground.

In this free-wheeling chat, Poornima Dore, Senior Program Manager at Tata Trusts, shares insights on what it’ll take for India to become a data-driven nation, and what gives her hope amidst formidable odds. Below are edited excerpts.

What influences the Tata Trusts’ approach to governance issues?

The Trusts have been working on development issues across the country for many years. Most of our work is a response to needs that have come up directly from the ground, or to issues of national importance. We have granular ground experience, and that’s what shapes our approach towards governance.

It’s important to note here that we are not trying to create a parallel framework to what the government is already doing. As one of the country’s largest philanthropies, we have the appetite to take risks. We are using [this advantage] to test new program designs, pilot them, and then scale them by collaborating with relevant partners on the ground.

One of the gaps we see is equipping decision makers with all the data they need in order to identify need areas and track the outcomes of policy interventions. This is critical for any government to deliver on its mandate.

That’s why one of our key focus areas now is building a culture of data-driven governance.

Since you brought up the role of data in governance, it’s a subject that has many myths attached to it.


For instance, does the government collect enough data to begin with?

The government does collect a fair amount of data as part of its regular tracking mechanisms, whether it’s through anganwaadi workers, the government healthcare system, local administrators, etc. However, data sharing between various government departments and triangulating all the data that is collected from different sources remains a problem, probably because it hasn’t been mandated. There is no centralized agency that is doing this work.

The other challenge is that a lot of government data is not real time. For certain parameters there is a dependency on census data, which is collected once in 10 years. Four or five years down the line, it’s very difficult for an administrator to rely on that data.

The other myth pertains to the quality of government data.

Yes. Is the data reliable? Is there a maker-checker concept? Those are valid questions. Having said that, I think we are still early in the game in India.

The biggest challenge right now is that we have a lot of data, but all that data is stuck in silos.

So you know the reality in bits and pieces but struggle to get the complete picture.

Are enough people in the government willing to use data for better governance? What’s your experience?

We have found that many top administrators are very interested. Whether they are district administrators, someone like Mr Aboobacker Siddique, former Deputy Commissioner of West Singbhoom; or even the political leadership, for instance Mr Srinivas Kesineni, the Member of Parliament from Vijayawada

Yes, the Vijayawada case study is legendary within SocialCops! We learned so much about how to empower change makers from that one exercise.

See, change is ultimately also a function of how excited the individual is.

If we can identify a few interested individuals and demonstrate [the power of data-driven governance] to them, even those who are cynical now will come on board.

Someone has to prove that it can be done. Once that happens, broader change will come partly because of optimism and partly because of the spirit of competition.

Today, awareness of what data can do is not a problem. People in the government know what the private sector is achieving through data analytics. The challenge is breaking down bureaucratic hurdles and creating a culture where data is central.

You touched upon the need to work with the right partners to build this culture. Could you elaborate?

Thank you, that’s a very important question. The scale and complexity of the work we do means that we need partners at every step to move the needle. First, we need a champion within the government or the bureaucracy who is willing to push a particular initiative and stay the course. They are probably the most important stakeholders, because at the end of the day the solutions we build are meant to empower them. If it’s a data-collection drive, they are the ones who need to come forward first and say ‘yes, we want this data, and we will use this data.’ Thereafter, it’s crucial to keep them engaged right through the process, whether it’s deciding the survey location, designing the survey, or formulating the questions.

Next, there’s the implementation partner. We need someone who understands the geography and the local community and can mobilize the youth in the area to actually collect the data.

The other important player is the technology partner. That’s the role SocialCops has been playing, and it’s a relationship we value a lot.

Once we decide what the goals of the survey are, the technology partner helps us host it on mobile devices, manage the back end, analyze the data, and finally visualize it such that decision maker can see exactly what he needs to.

Finally, we need to embed a regular monitoring process within the administration, so that the data we collect and the dashboards we build are actually used to target and reach the right beneficiaries.

With every new survey, we need to refine and customize each of these processes. So we invest a lot in our partners to make sure that they are [up to speed].

Earlier, you talked about overcoming cynicism within the administration. I was wondering about citizen cynicism. How big a challenge is that?

The citizen is the ‘customer’ for us. If they don’t feel the benefit, then all this work is pointless. The problem is, there are multiple data collection drives happening through the year. Different government agencies are collecting data for different programs. Each time a new scheme is announced, there’s a survey. That’s where the cynicism creeps in; people start thinking, “what is going to come out of all this?” Actually, that strengthens the case for better sharing and triangulation of data across government departments, so that citizens don’t have to go through surveys again and again.

So, how do you enthuse a villager to participate in yet another government survey?

First, we make the survey process as simple as possible. We keep the questionnaires short, so that we only capture the most important data and don’t take up too much time.

More important, we emphasize people’s participation. Before rolling out a survey in a village, we conduct focus group discussions involving the villagers, where they decide the village’s developmental needs. During the survey, we make sure that the volunteers collecting the data include locals.

Integrating the citizens also helps because, when we take the data to the administration, they know that it’s authentic. Hopefully, that motivates them to move from the standard “10% annual increase” in allocations for various activities to a more nuanced, data-driven approach.

In Noamundi, for instance, we now have data at the beneficiary level for certain pension schemes. The data tells us which citizens are eligible for these schemes but are not accessing them. Now that the administration has this data, they can proactively reach out to the eligible citizens and bring them under coverage. They look at the dashboard, and they know they have to turn the reds into greens. [This sets in motion a virtuous cycle] — citizens see the benefit of sharing data, and next time there’s a survey, they are more enthusiastic.

Finally, a key long-term goal for the government should be to see how data collected from citizens can be made publicly available at least at an aggregate level. Maybe there could be logins created for panchayats, so that citizens could report problems with the data or update the data on their own.

Okay, let’s step back for a moment. Tell us some of your favorite work stories.

{Long pause} Some of the most heartwarming stories come from the volunteers collecting the data for our surveys … the change we see in them from the time they start working with us to the end of the exercise is amazing. Many of them are school dropouts. They aren’t comfortable using mobile devices. They don’t even know their own villages that well.

By the end, they come to us and tell us that they feel so much more empowered to do something for their village. This is a place where they have lived all their lives, but it’s only when they work on a survey that they get to truly experience every aspect of it beyond just their own personal circles.

Their confidence levels and their respect within the community also go up because they are seen as smart people who can handle technology and get things done. They pick up the ability to complete complex tasks within hard deadlines. In a sense, they become fully trained as project managers.

The other kind of high comes from the reaction of the administrators. In Noamundi and Chandrapur, for example, the district collectors were very excited when we shared the data with them. They got curious. They started asking very pointed questions and demanding new data points. It tells you that the end user for whom you are building all this is in fact interested. It motivates you to do whatever it takes to give them what they need.

We recently presented our work on data-driven governance to our board of trustees, including Mr Ratan Tata. There was a lot of serious interest about how citizens can use data better. It was very heartening that such senior leaders, who know how challenging this kind of work can be, feel that it has potential. It gives you hope.

Finally, a risk with people who work with data all the time is that they start believing in the absolute power of data. How important do you think it is to marry data with intuition?

All data must have a context. Without context, you are basically using data to suit your own agenda [and not to get to the truth]. If data is moving in a certain [exaggerated] way because the area in question is a tribal belt, or because it is a very remote village, you need to be able to build in that context into your understanding of the data. What has helped us [to be mindful of context] is our clear focus on using data to drive service delivery and not to run fancy analytics or regressions just for the sake of it.

Ultimately, data is just an aid. You cannot take away your intelligence and intuition while interpreting it.